[Documentation] Use when generating, smart-merging or refactoring the root AI context (CLAUDE.md, AGENTS.md) and its Claude/Codex projections. Config file only: project-config.
The English summary is being prepared. Showing the original description.
[Content] Use when a workflow step or the user asks for course material: Bloom objectives, modules, lessons, exercises and assessments.
The English summary is being prepared. Showing the original description.
See how to installReview the instructions the agent will receive before you install.
Goal: Build a learner-ready course with Bloom-aligned objectives, progressive modules, lessons, practice, and assessments so learners achieve and demonstrate intended outcomes.
Summary:
TaskCreate tasks before work; map every objective to Bloom; cite every factual claim inline [N]; use the enforced template; write docs/knowledge/courses/{descriptive-slug}.md; no modes/flags specified.Workflow:
Key Rules:
.claude/templates/course-outline-template.md.[N] citation mapped to a Resources & References row (Title, URL, Author/Org, Date, Tier). Factual claims rest on 2+ independent sources; a single-source claim is marked unverified and held below 60% confidence. Exercises, assessment items and constructed examples are course design, not factual claims.Critical thinking: Be skeptical; reason sequentially; trace evidence for every claim; state confidence percentages; ideas >80%.
Classify course type, learner level, delivery mode, and constraints before choosing depth, examples, exercises, or assessment formats; ask for missing inputs or state assumptions.
| Level | Verb Examples | Assessment Type |
|---|---|---|
| Remember | Define, list, recall, identify | Multiple choice, fill-in-blank |
| Understand | Explain, describe, summarize, interpret | Short answer, paraphrase |
| Apply | Use, implement, solve, demonstrate | Problem sets, exercises |
| Analyze | Compare, contrast, examine, differentiate | Case studies, analysis papers |
| Evaluate | Judge, critique, assess, justify | Debates, reviews, peer assessment |
| Create | Design, construct, produce, develop | Projects, portfolios, presentations |
Gather from user or research (in a research run, read the evidence base tmp/research/_evidence-{slug}.md that web-research --chain=deep-dive returned and cite its sources): Target audience — who + background; Prerequisites — prior knowledge; Duration — total hours/weeks; Desired outcomes — learner capabilities after course.
Assign a Bloom's level to each desired outcome:
Organize 3-8 modules per course:
For each lesson, provide: (1) Duration — estimated time; (2) Concept — core idea in 1-2 sentences; (3) Explanation — theory, context, why it matters; (4) Examples — 2-3 real-world illustrations; (5) Exercise — hands-on practice activity; (6) Assessment — how to verify learning.
Per module: Knowledge check — 3-5 questions covering key concepts; questions align with module's Bloom's level. Final: Comprehensive assessment — covers all modules; Mix of Bloom's levels — at least 1 question per level taught.
Verify objective/activity/assessment alignment, evidence for content claims, labelled assumptions, progressive difficulty, and prerequisite chain; state confidence.
Write to docs/knowledge/courses/{descriptive-slug}.md using enforced template from .claude/templates/course-outline-template.md. List every cited source in Resources & References with Title, URL, Author/Org, Date and Tier; every [N] resolves to a row and every row is cited.
[IMPORTANT] Create small tasks with
TaskCreatebefore work, including a final review task.
IMPORTANT MUST ATTENTION Goal: Build a learner-ready course with Bloom-aligned objectives, progressive modules, lessons, practice, and assessments so learners achieve and demonstrate intended outcomes. IMPORTANT MUST ATTENTION Course flow: Phase 0 detect context; (1) define scope; (2) map every objective to Bloom; (3) structure 3-8 modules with 2-5 lessons and prerequisite links; (4) develop each lesson's duration, concept, explanation, examples, exercise, and assessment; (5) create 3-5-question module checks plus a comprehensive final with every taught Bloom level; (6) review alignment, evidence, assumptions, progression, and prerequisites.
IMPORTANT MUST ATTENTION Gate/output: create TaskCreate tasks before work and a final review task; cite every factual claim inline [N] mapped to a Resources & References row (Title, URL, Author/Org, Date, Tier), with a single-source claim marked unverified below 60% confidence; use enforced .claude/templates/course-outline-template.md; write docs/knowledge/courses/{descriptive-slug}.md; no modes/flags specified.
IMPORTANT MUST ATTENTION use user/research evidence, label assumptions, and state confidence; NEVER fabricate course facts.
Anti-Rationalization:
| Evasion | Rebuttal |
|---|---|
| "Course is simple" | Wrong assumptions waste time; apply the full flow. |
| "This fact is common knowledge" | Cite it [N] to a Resources & References row; with one source, mark it unverified below 60%. |
| "Just do it" | Create task tracking first; skip depth only when justified, never skip tracking. |
[TASK-PLANNING] MUST ATTENTION Before acting, analyze scope; create small TaskCreate tasks, cite every factual claim inline [N] to a Resources & References row, and complete a final quality review.
No reviews yet. Tell others what you thought.
No login needed. Share how it worked for you, ask a question, or note where it did not run.
No comments yet. Be the first to write one.
[Documentation] Use when generating, smart-merging or refactoring the root AI context (CLAUDE.md, AGENTS.md) and its Claude/Codex projections. Config file only: project-config.
The English summary is being prepared. Showing the original description.
[Code Quality] Use when a workflow step or the user asks for AI code or plan review: LLM calls, prompts, agents, RAG, evals and guardrails. --mode={code|plan}; --report-only is read-only.
The English summary is being prepared. Showing the original description.
[Architecture] Use when a workflow step or the user asks for architecture design, compliance review, scalability grading or whole-project audit via --mode={design|review|scalability|full}.
The English summary is being prepared. Showing the original description.
[Content] Use when a workflow step or the user asks for product/feature ideation with ranked opportunities and validation tests. --mode={roadmap|scope}.
The English summary is being prepared. Showing the original description.
[Content] Use when a workflow step or the user asks for business viability: Business Model Canvas, financials, risks and go-to-market recommendation.
The English summary is being prepared. Showing the original description.
[Code Quality] Use when a workflow step or the user asks for a focused code-change review: staged, unstaged, branch diff or simple pull request. --fix-loop fixes and re-reviews. Complex pull request: workflow-review-changes; rationale: why-review.
The English summary is being prepared. Showing the original description.